2 Clarke Drive
Cranbury, NJ 08512
© 2022 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
Although there has been significant progress in the field of geriatric oncology, experts say there is still a pressing need to increase clinical trial participation by older patients and to improve the predictive value of assessment tools.
Elisabeth A. Quoix, MD
In an era of personalized care, advancements in geriatric oncology have resulted in tailored therapeutic strategies, extending the benefits of checkpoint immunotherapy and other novel agents to older patients. Although there has been significant progress in the field, experts say there is still a pressing need to increase clinical trial participation by older patients and to improve the predictive value of assessment tools.
Traditionally, older patients were considered too frail to undergo surgery or chemotherapy, but views on therapies for this population have been shifting since at least the mid-1980s.1 Today, an evolving understanding of comorbidities and other clinical factors has enabled older patients to be considered for novel and established regimens.
This year’s American Society of Clinical Oncology Annual Meeting (ASCO 2019) provided several signs of encouragement for geriatric oncology specialists. Among many abstracts that included data about older patients, several late-breaking presentations are of particular interest, according to a review in the Journal of Geriatric Oncology (Table 1).2-8
“The enthusiastic momentum of geriatric oncology is evident by an increasing number of abstracts presented at the ASCO annual meeting throughout multiple tracks,” Sheikh and colleagues wrote. “Overall, these trials advance the ongoing effort to better define the optimum care for older adults with cancer.”2
Elisabeth A. Quoix, MD, who leads clinical trials for patients with non—small cell lung cancer (NSCLC) who are 70 years and older, received the B.J. Kennedy Award for Scientific Excellence in Geriatric Oncology at ASCO 2019. “The main take-home message is that there is no more room for nihilism regarding the treatment of elderly patients,” Quoix, a professor of pneumonology in the Department of Chest Diseases of the University Hospital of Strasbourg, France, said in delivering the award lecture about NSCLC studies.9
Clinical Trial Shortcomings
Despite her upbeat message, Quoix noted that studies to assess therapeutic options for older patients are needed. Although her comments were made in the context of NSCLC research, they have been echoed by experts in other malignancies.
“There is usually no significant survival difference between the elderly and their younger counterparts in age-unspecified studies,” Quoix said. “Elderly patients included in the age-unspecified trials are not as old as in specific studies, and they are highly selected. Adverse events [AEs] are more frequent in the age-unspecified trials. This has been shown in several studies.
“Thus, there is a major limitation of generalizing the conclusions of an age-unspecified trial to all elderly cancer patients, and this is the reason for which [studies dedicated] to older patients are mandatory,” Quoix said.
Although some trials are now being designed for older patients, age disparities in clinical trial populations “are pervasive, worsening, and associated with industry sponsorship,” according to study findings presented at ASCO 2019.10
Investigators analyzed 302 trials listed on ClinicalTrials.gov with posted results that evaluated therapeutic interventions in breast, prostate, colorectal, and lung cancers. They found a deviation between the median ages of trial participants and the population by disease site in Surveillance, Epidemiology and End Results (SEER) program data in several comparative categories.
The studies were compared for several factors, including performance status, the use of a molecular biomarker as criteria, and sponsorship. Although there was an age gap in trials sponsored by cooperative groups, there were a greater number of industry- sponsored studies with age disparities, and the difference was statistically significant (P = .002).
Those findings are in keeping with prior studies that have documented the underrepresentation of older patients in clinical trials.
“That’s really the biggest challenge in geriatric cancer care: A study is being reported. The median age of patients in that study is 60, but when you look at the patients you’re treating for the same condition, they’re mostly in their late 70s,” said Efrat Dotan, MD, who chairs the Older Adult Oncology Guidelines for the National Comprehensive Cancer Network (NCCN). “It’s hard for the clinician to know how the trial results apply. Do you use a dose adjustment? Do you go by what was done in the study? How should comorbidities or medications affect treatment? Most trials don’t address any of those real-world questions.”
Comprehensive geriatric assessments (CGAs) that go beyond chronological age to include comorbidities and other clinical factors are one mechanism that the NCCN guidelines recommend for evaluating older patients for therapy.11
“Numerous studies have shown that performing assessments before treatment improves outcomes, usually quality-of-life outcomes, in geriatric patients as a whole,” said Dotan, an associate professor in the Department of Hematology/Oncology at Fox Chase Cancer Center in Philadelphia, Pennsylvania. “Now various studies are taking it further and developing disease-specific tools or tools that help you predict chemotherapy toxicity or tools that help you use an assessment to choose between different treatments.”
Geriatric cancer is a new enough research specialty that many fundamental questions remain unanswered, including what constitutes a geriatric patient. Many published studies define geriatric patients entirely by their age, usually ≥65 years.
By this standard, geriatric patients already constitute the majority of all patients with cancer in the United States. An analysis of SEER data from 2007 to 2011 found that the average patient age at cancer diagnosis was 66 and that that number was expected to increase significantly as the nation’s population ages.12 Approximately 60% of all people who currently have cancer are ≥65.13
However, chronological age isn’t the only factor in identifying which older patients might be ill-served by standards of care evaluated in younger populations. Experts say at least as much consideration should be given to physical age. For example, a 60-year-old with type 2 diabetes and a history of heart disease is, in most respects, more of a geriatric patient than a 70-year-old fitness buff.
A CGA takes such factors into consideration through evaluation in 5 main areas: comorbidities, cognitive function, nutritional status, polypharmacy, and medication review, according to the NCCN guidelines.11
Quoix said that although age-related changes affect treatment, more than 50% of elderly patients consider an increase in survival as the main objective of therapy. Clinical factors that affect pharmacodynamics are: reduced total body water; an increase of body fat; reduced renal, hepatic, or hematopoietic function reserves; and gastrointestinal changes. Medical problems that can affect therapy are comorbidities, polymedication, and geriatric syndromes (Table 2).9
Although older patients with cancer may have been excluded from certain treatments in the past, immune checkpoint inhibitor (ICI) therapy has opened up a potential new option for all frail patients, including the elderly population. In urothelial carcinoma, for example, pembrolizumab (Keytruda) and atezolizumab (Tecentriq) are approved for patients with advanced or metastatic disease who are not eligible for any platinum-containing therapy, regardless of PD-L1 status.14
“How many total patients are there who are candidates for ICIs but would not have been candidates for any previously approved treatments for their tumor types? I don’t think we know, but it’s a significant population, and it keeps growing as ICIs keep winning new approvals,” said Rawad Elias, MD, a hematologist/oncologist at Hartford Hospital who has studied immunotherapy in older adults. “Much more significant is the number of geriatric patients who would have undergone some form of chemotherapy who now have a better option.”
In light of the relative shortage of geriatric patients with cancer in most ICI trials, there have been some questions about both the efficacy and toxicity of such medications in this group. Findings from several recent studies indicate that positive and negative outcomes are similar among age groups. In a 2019 meta-analysis, Kasherman et al pooled data from 19 randomized trials of ICI monotherapy in patients with advanced cancer and found that the overall survival (OS) benefit was not significantly related to patient age. The hazard ratios were 0.73 for those ≥65 years and 0.79 for those <65 years (P interaction = .27). Findings were similar at older age cutoffs of 70 years, with data from 2 studies indicating hazard ratios of 0.93 for those ≥70 years (P interaction = .91) and 0.95 for patients <70 years (P interaction = .91). This also was the case for 3 studies with data from patients ≥75 years, which showed hazard ratios of 0.75 for those ≥75 years and 0.61 for those <75 years (P interaction = 0.72).15
There was, however, an important caveat: Although the meta-analysis contained more than 10,000 people who were over the chronological age of 65, it did not include many patients with comorbidities and other health problems associated with advanced physical age. All but a handful of the patients entered their respective trials with an ECOG performance status of either 0 or 1.15
The conclusions of that study correspond with those of a 2018 meta-analysis. Investigators pooled data from 9 ICI trials and evaluated the findings using a random/mixed effects model to account for heterogeneity between studies. The overall estimated random-effects hazard ratio for death was 0.64 (95% CI, 0.54-0.76) in patients ≥65 years versus 0.68 (95% CI, 0.61-0.75) in patients <65 years. The overall estimated random-effects for hazard ratio for progression was 0.74 (95% CI, 0.60-0.92) in patients ≥65 years versus 0.73 (95% CI, 0.61-0.88) in patients <65 years.16
Neither of the meta-analyses reported on AEs, but findings from several smaller studies suggest that geriatric patients experience a similar incidence and severity of AEs as do younger patients treated with ICIs.
A 2019 study from Muchnik et al reported ICI toxicity data from 75 patients with NSCLC, aged 70 years or older, who were treated at the Wilmot Cancer Institute at the University of Rochester in New York. Unlike clinical trials restricted to populations with more favorable characteristics, these realworld patients were not unusually healthy: 53% had a Charlson Comorbidity Index (CCI) of 3 or higher and 49% had an ECOG performance status of 2 or higher. Overall, 37% of patients experienced an immune-related AE (irAE) of any grade, and 8% were grade 3 or higher. Of those who discontinued ICIs, 15% discontinued because of irAEs. Of those who experienced irAEs, 64% received glucocorticoids. Toxicity generally did not differ by age, CCI, or ECOG score.17
A 2018 study by Silva et al included 106 patients aged 65 and older who were treated for common stage IV cancers in clinics around Rio de Janeiro, Brazil. All-grade AEs occurred in 55% of patients and led 13 to discontinue treatment. There were 21 irAEs, the most common of which were thyroiditis, rash, and pruritus. There were 4 severe irAEs: colitis, antibody-enhanced anemia, fulminant hepatitis and polymyositis. The only predictive variable for AE was patient frailty (OR, 3.03; 95% CI 1.36-6.74; P = .006).18
Notably, the frequency of AEs is not the only factor in tolerability.
“The other big concern is how quickly and how well patients recover from adverse events, and geriatric patients certainly don’t recover as quickly or as well as [younger] patients do,” Elias said. “That doesn’t mean they don’t tolerate ICIs better than they tolerate systemic chemotherapy, but it does mean they necessarily tolerate ICI toxicity as well as younger patients do.”
Elias noted a meta-analysis exploring fatal toxicities associated with ICIs. Although these events are rare with ICIs, patients who died of toxic effects tended to be older than those without such AEs (70 vs 62 years; P = .009).19 An analysis of the FDA Adverse Event Reporting System showed the prevalence of AEs was higher in adults 65 years or older than in those younger than 65; additionally, the prevalence of death was higher in older adults who suffered a toxicity after ipilimumab (Yervoy) monotherapy and nivolumab (Opdivo) plus ipilimumab combination therapy than it was for younger patients. No significant difference across age group was observed with PD-1/PD-L1 ICIs.20
Efforts to Decrease Toxicity
Given the prevalence of combination regimens and the difficulty that frail patients can have in tolerating such therapy, investigators have begun comparing strategies in clinical trials that focus exclusively on older patients.
In the phase III GO2 trial conducted in the United Kingdom, investigators evaluated alternative regimens for elderly patients with advanced gastroesophageal cancer who were considered unfit for fulldose 3-drug standard care with epirubicin/ oxaliplatin/capecitabine. To date, the study is the largest randomized controlled clinical trial specifically designed for frail elderly patients in this population, investigators reported at ASCO 2019.21
Patients were randomized to receive 100% (level A), 80% (level B), or 60% (level C) of the full dose of oxaliplatin (130 mg/m2 on day 1 of a 21-day cycle) plus capecitabine (625 mg/ m2 twice daily) until progression. They also underwent CGAs for 9 domains: weight loss, mobility, falls, neuropsychiatric, function, social, mood, fatigue, and polypharmacy. The CGAs were then scored based on the number of domains in which patients were identified as impaired: not frail, 0; mildly frail, 1 to 2; and severely frail ≥3.
Of the 512 patients who participated in the study, the median age was 76 years (range, 51-96) and 58% were considered severely frail. The findings showed that both of the lower doses were associated with noninferior progression-free survival (PFS) compared with level A; the hazard ratio for level B versus A was 1.09 (95% CI, 0.89-1.32) and for level C versus A, 1.10 (95% CI, 0.90-1.33). The median OS in the 3 dosing levels was, respectively, 7.5, 6.7, and 7.6 months.
The patients who received the lowest dose tested experienced less toxicity and also fared better on an overall treatment utility, a metric that the study investigators devised to measure clinical and patient-reported outcomes, as well as on toxicity levels and quality-of-life standards. Moreover, there was no subgroup group that benefited more from the higher dose levels.21 Investigators are now working on personalizing dose selection based on CGA ratings.
In the phase III A041202 study, the Alliance for Clinical Trials in Oncology compared ibrutinib (Imbruvica) either as monotherapy or in combination with rituximab (Rituxan) versus standard-of-care chemoimmunotherapy with bendamustine plus rituximab in patients ≥65 years with untreated chronic lymphocytic leukemia (CLL).22
The study enrolled 547 patients with a median age of 71 years (range, 65-89), most (97%) with ECOG performance scores of 0 or 1. At 2 years, the estimated PFS rate among those in the primary analysis was 87% (95% CI, 81%-92%) with ibrutinib monotherapy (n = 178), 88% (95% CI, 81%-92%) with ibrutinib combination therapy (n = 170), and 74% (95% CI, 66%-80%) with the bendamustine regimen (n = 176).
There was no significant difference in PFS between the 2 ibrutinib arms of the study, although there were when those arms were compared with the bendamustine combination. The hazard ratio for disease progression or death was 0.39 (95% CI, 0.26- 0.58; 1-sided P <.001) for the ibrutinib-only and 0.38 (95% CI, 0.25-0.59; 1-sided P <.001) for the ibrutinib combination arms versus the bendamustine cohort. However, there was no significant difference in OS between the 3 treatment groups.
Toxicity profiles also differed depending on the regimen. The rate of grade 3, 4, or 5 hematologic AEs was lower in the ibrutinib monotherapy or combination arms (41% and 39%, respectively) than for the bendamustine-containing arm (61%), whereas nonhematologic AEs of this severity were higher with ibrutinib therapy (74% for each arm) than for the bendamustine cohort (63%).
Another Step for Assessments
There are many ways in which CGAs can help improve care for older patients, including uncovering problems not identified in routine oncologic care, predicting AEs, influencing treatment decisions, and estimating life expectancy.11
Findings from several recent studies demonstrate that CGAs may have value in a number of tumor types:
Nevertheless, other study results show that CGAs are not significantly associated with outcomes. In the A041202 CLL study, CGAs were not statistically significant among the 3 treatment groups.22 In her presentation, Quoix cited several examples in which CGAs were not predictive of NSCLC outcomes.9
Moving forward, geriatric oncologists will seek to refine the use of these evaluations.
“A lot of the early research in geriatric oncology, which is still a relatively new field, focused on risk prediction. It found ways to use geriatric assessment characteristics to predict either toxicity or survival associated with particular treatments, and that work has built a basis for understanding how geriatric assessment can improve cancer care,” said Melisa L. Wong, MD, MAS, an assistant professor in the Division of Hematology/Oncology at the University of California, San Francisco.
“Now there are many ongoing studies that are looking at how to use geriatric assessments to better guide care by considering factors beyond toxicity, factors such as functional status or cognition or quality of life as outcomes.”